A pseudo-DNS method for the simulation of incompressible fluid flows with instabilities at different scales

In this work, a new model for the analysis of incompressible fluid flows with massive instabilities at different scales is presented. It relies on resolving all the instabilities at all scales without any additional model, i.e., following the direct numerical simulation style. Nevertheless, the computation is carried out at two levels or scales, termed the coarse and the fine. The fine-scale simulation is performed on representative volume elements providing the homogenized stress tensor as a function of several dimensionless numbers characterizing the flow. Consequently, the effect of the fine-scale instabilities is transferred to the coarse level as a homogenized stress tensor, a procedure inspired by standard multi-scale methods used in solids. The present proposal introduces a new way for the treatment of the flow at the fine scale, simulating not only the coarse scale but also the fine scale with all the necessary detail, but without incurring in the excessive computational cost of the classical DNS. Another interesting aspect of the present proposal is the use of a Lagrangian formulation for convecting the eddies simulated on the fine mesh through the coarse domain. Several examples showing the potentiality of this methodology for the simulation of homogeneous flows are presented.

[1]  A. Morozov From chaos to order in active fluids , 2017, Science.

[2]  Eugenio Oñate,et al.  Advances in the particle finite element method (PFEM) for solving coupled problems in engineering , 2010 .

[3]  Sergio R. Idelsohn,et al.  A second-order in time and space particle-based method to solve flow problems on arbitrary meshes , 2019, J. Comput. Phys..

[4]  Charles Meneveau Turbulence: Subgrid-Scale Modeling , 2010, Scholarpedia.

[5]  D. Lilly,et al.  A proposed modification of the Germano subgrid‐scale closure method , 1992 .

[6]  C. G. Speziale On nonlinear K-l and K-ε models of turbulence , 1987, Journal of Fluid Mechanics.

[7]  Robert McDougall Kerr,et al.  Higher-order derivative correlations and the alignment of small-scale structures in isotropic numerical turbulence , 1983, Journal of Fluid Mechanics.

[8]  J. Deardorff A numerical study of three-dimensional turbulent channel flow at large Reynolds numbers , 1970, Journal of Fluid Mechanics.

[9]  T. Gotoh,et al.  Pressure spectrum in homogeneous turbulence. , 2001, Physical review letters.

[10]  Yukio Kaneda,et al.  High-resolution direct numerical simulation of turbulence , 2006 .

[11]  Eugenio Oñate,et al.  The particle finite element method: a powerful tool to solve incompressible flows with free‐surfaces and breaking waves , 2004 .

[12]  Chandra S. Martha,et al.  Large eddy simulations of 2-D and 3-D spatially developing mixing layers , 2011 .

[13]  Sergio Idelsohn,et al.  A fast and accurate method to solve the incompressible Navier‐Stokes equations , 2013 .

[14]  Kai Schneider,et al.  Coherent vortices in high resolution direct numerical simulation of homogeneous isotropic turbulence: A wavelet viewpoint , 2007 .

[15]  Eugenio Oñate,et al.  Particle-Based Methods , 2011 .

[16]  Hans Edelmann,et al.  Vier Woodbury-Formeln hergeleitet aus dem Variablentausch einer speziellen Matrix , 1976 .

[17]  Steven A. Orszag,et al.  Numerical Methods for the Simulation of Turbulence , 1969 .

[18]  Shiyi Chen,et al.  On statistical correlations between velocity increments and locally averaged dissipation in homogeneous turbulence , 1993 .

[19]  Tohru Nakano,et al.  Velocity field statistics in homogeneous steady turbulence obtained using a high-resolution direct numerical simulation , 2002 .

[20]  C. K. Chan,et al.  Large eddy simulation of mixing layer , 2004 .

[21]  A. Vincent,et al.  The spatial structure and statistical properties of homogeneous turbulence , 1991, Journal of Fluid Mechanics.

[22]  J. M. Giménez Enlarging time-steps for solving one and two phase flows using the particle finite element method , 2015 .

[23]  W. Rodi A new algebraic relation for calculating the Reynolds stresses , 1976 .

[24]  A. Hellsten,et al.  New Advanced k-w Turbulence Model for High-Lift Aerodynamics , 2004 .

[25]  P. Sagaut,et al.  Large Eddy Simulation for Compressible Flows , 2009 .

[26]  A. N. Kolmogorov Equations of turbulent motion in an incompressible fluid , 1941 .

[27]  T. Hughes,et al.  Large Eddy Simulation and the variational multiscale method , 2000 .

[28]  Juan M. Gimenez,et al.  Evaluating the performance of the particle finite element method in parallel architectures , 2014, Computational Particle Mechanics.

[29]  Eric D. Siggia,et al.  Numerical study of small-scale intermittency in three-dimensional turbulence , 1981, Journal of Fluid Mechanics.

[30]  Ugo Galvanetto,et al.  Multiscale modeling in solid mechanics : computational approaches , 2009 .

[31]  P. Moin,et al.  A dynamic subgrid‐scale eddy viscosity model , 1990 .

[32]  Javier Jiménez,et al.  The structure of intense vorticity in isotropic turbulence , 1993, Journal of Fluid Mechanics.

[33]  P. Blanco,et al.  Homogenization of the Navier-Stokes equations by means of the Multi-scale Virtual Power Principle , 2017 .

[34]  J. Smagorinsky,et al.  GENERAL CIRCULATION EXPERIMENTS WITH THE PRIMITIVE EQUATIONS , 1963 .

[35]  A. Huespe,et al.  Multi-scale (FE2) analysis of material failure in cement/aggregate-type composite structures , 2014 .

[36]  J. Marti,et al.  An explicit–implicit finite element model for the numerical solution of incompressible Navier–Stokes equations on moving grids , 2019, Computer Methods in Applied Mechanics and Engineering.

[37]  B. Launder,et al.  The numerical computation of turbulent flows , 1990 .

[38]  K. Abdol-Hamid Development and Documentation of kL-Based Linear, Nonlinear, and Full Reynolds Stress Turbulence Models , 2018 .

[39]  Y. Kaneda,et al.  High Resolution DNS of Incompressible Homogeneous Forced Turbulence —Time Dependence of the Statistics— , 2003 .

[40]  I. Wygnanski,et al.  The forced mixing layer between parallel streams , 1982, Journal of Fluid Mechanics.